CT Batch Standardisation Pre-processing Software

Development Background In the research of medical imaging artificial intelligence, raw CT data often contains invalid Hounsfield Unit (HU) values (such as<-1024 or>3071), which affects the stability of model training; At the same time, different devices or scanning parameters result in inconsistent image display effects, increasing the burden on doctors to read the images. Although open source libraries such as nibabel and numpy support data processing, they typically require users to have programming skills and manually implement operations such as HU cropping and window width and level configuration.

At present, there is a lack of a lightweight tool for CT image standardization that is programming free, environment free, and dedicated to ordinary users. To this end, we have developed this software that encapsulates professional preprocessing logic into independent executable programs, providing a one click operation experience and significantly reducing the threshold for use. Software functions and features 1) Zero environment dependency: No need to install Python or any third-party libraries, just unzip and use; 2) Keep the original HU: Only crop extreme invalid values without changing the distribution of the original HU within the valid range, ensuring the authenticity of the AI model input; 3) Support default window settings: Write WL/WC through the NIfTI file header (cal_in/cal_max field), compatible with mainstream medical imaging visualization software; 4) Built in common window reference: The interface provides typical parameters such as lung window, soft tissue window, bone window, brain window, etc., making it convenient for users to quickly choose; 5) Batch efficient processing: Automatically traverse all. nii.gz files in the input directory and output standardized results with the same name; 6) Graphic interface: Simple and intuitive operation, even non-technical users can easily get started.